Postgres Pro
Downloads
Home   >   mailing lists

Re: probably cause (and fix) for floating-point assist faults on itanium - Mailing list pgsql-performance

From Claudio Freire
Subject Re: probably cause (and fix) for floating-point assist faults on itanium
Date
Msg-id CAGTBQpYjs8GHBk1mf__dYVH=4g7ZK2jHrN7ubNYJtYX+47OzNg@mail.gmail.com
Whole thread Raw
In response to probably cause (and fix) for floating-point assist faults on itanium  (Greg Matthews <gregory.a.matthews@nasa.gov>)
Responses Re: probably cause (and fix) for floating-point assist faults on itanium
List pgsql-performance
Tree view 
On Thu, Nov 17, 2011 at 10:07 PM, Greg Matthews
<gregory.a.matthews@nasa.gov> wrote:
>        if (smoothed_alloc <= (float) recent_alloc)
>                smoothed_alloc = recent_alloc;
>        else if (smoothed_alloc >= 0.00001)
>                smoothed_alloc += ((float) recent_alloc - smoothed_alloc) /
>                        smoothing_samples;
>

I don't think that logic is sound.

Rather,

       if (smoothed_alloc <= (float) recent_alloc) {
               smoothed_alloc = recent_alloc;
       } else {
               if (smoothed_alloc < 0.000001)
                   smoothed_alloc = 0;
               smoothed_alloc += ((float) recent_alloc - smoothed_alloc) /
                       smoothing_samples;
       }

pgsql-performance by date:

Previous
From: MirrorX
Date:
Subject: Re: index usage for min() vs. "order by asc limit 1"
Next
From: Tom Lane
Date:
Subject: Re: probably cause (and fix) for floating-point assist faults on itanium